%%capture
!pip install geopandas
!pip install lxml

import zipfile
import requests
import io
import numpy as np
import pandas as pd
import geopandas as gpd
import matplotlib.pyplot as plt
import seaborn as sns
from IPython.display import Markdown
from wordcloud import WordCloud

import plotly.express as px

url = 'https://github.com/erikbozik/Earthquakes/raw/main/hazards_geojson.zip'

data = requests.get(url)

with zipfile.ZipFile(io.BytesIO(data.content)) as zip_ref:
    with zip_ref.open('hazards_geojson/earthquakes.geojson', 'r') as file:
        earthquakes = gpd.read_file(file)
    with zip_ref.open('hazards_geojson/volEvents.geojson', 'r') as file:
        volEvents = gpd.read_file(file)
    with zip_ref.open('hazards_geojson/tsunamiEvents.geojson', 'r') as file:
        tsunamiEvents = gpd.read_file(file)
    with zip_ref.open('hazards_geojson/tsunami_runups.geojson', 'r') as file:
        tsunami_runups = gpd.read_file(file)
    print('succesfully loaded all data')

succesfully loaded all data

# # Convert GeoJSON to CSV
# earthquakes.to_csv('earthquakes.csv', index=False)
# volEvents.to_csv('volEvents.csv', index=False)
# tsunamiEvents.to_csv('tsunamiEvents.csv', index=False)
# tsunami_runups.to_csv('tsunami_runups.csv', index=False)

# from google.colab import files

# # Download CSV files
# files.download('earthquakes.csv')
# files.download('volEvents.csv')
# files.download('tsunamiEvents.csv')
# files.download('tsunami_runups.csv')

earthquakes1 = earthquakes.copy(deep=True).query("`magnitude`.notnull() and `deaths`.notnull()")
def deaths_parse(value):
    if value is not np.nan:
        index = value.find('(')
        key, value = value[: index].strip(), value[index + 2:]
        ran = ()
        for i in value.split():
             try:
                 ran += int(i),
             except ValueError:
                 pass
        if len(ran) != 2:
            out = (ran[0], '>')
        else:
            out = (ran[0], ran[1])
        return out
    return value


earthquakes1['deaths'] = earthquakes1['deaths'].apply(deaths_parse)
earthquakes1['magnitude'] = earthquakes1['magnitude'].astype(float)
earthquakes1 = earthquakes1.sort_values(by='deaths')
earthquakes1['deaths'] = earthquakes1['deaths'].apply(lambda x: f'{x[0]} - {x[1]}' if x[1] != '>' else f'{x[0]}>')

figure, axes = plt.subplots(figsize=(15, 10))
e_death_scatter = sns.boxplot(data=earthquakes1, x='deaths', y='magnitude')
e_death_scatter.set_ylabel('Magnitúda')
e_death_scatter.set_xlabel('Počet úmrtí')
e_death_scatter.set_title('Všetky zemetrasenia')
pass

figure, axes = plt.subplots(figsize=(15, 10))
earthquakes1['magnitude'] = earthquakes1['magnitude'].astype(float)


e_death_violin = sns.violinplot(data=earthquakes1, x='deaths', y='magnitude')

low_deaths_max_mag = earthquakes1.query("`deaths` == '1 - 50'")['magnitude'].max()
e_death_violin.set_title('Všetky zemetrasenia')
e_death_violin.set_ylabel('Magnitúda')
e_death_violin.set_xlabel('Počet úmrtí')
pass

# countries data
countries = gpd.read_file(gpd.datasets.get_path('naturalearth_lowres'))
country_areas = countries['geometry'].to_crs({'proj':'cea'}).area  / 1e6
countries['area'] = country_areas
density = (countries['pop_est']
           / countries['area'])
countries['pop_dens'] = density
display(countries.head())
pass

C:\Users\varga\AppData\Local\Temp\ipykernel_41628\2896388518.py:2: FutureWarning: The geopandas.dataset module is deprecated and will be removed in GeoPandas 1.0. You can get the original 'naturalearth_lowres' data from https://www.naturalearthdata.com/downloads/110m-cultural-vectors/.
  countries = gpd.read_file(gpd.datasets.get_path('naturalearth_lowres'))

high_magnitude_low_deaths = earthquakes1.copy(deep=True).query("`magnitude` > 7.5 and `deaths` == '1 - 50' and year > 1950")

main = px.choropleth(
    countries,
    locations=countries['iso_a3'],
    color='pop_dens',
    color_continuous_scale='Blues',
    range_color=(0, 250),
    projection='mollweide',
    title='Mapa nízkej úmrtnosti a nízkej hustoty obyvateľstva po roku 1950'
)
fig = px.scatter_geo(
    high_magnitude_low_deaths,
    lat=high_magnitude_low_deaths.geometry.y,
    lon=high_magnitude_low_deaths.geometry.x,
    hover_name="location",
    hover_data=['magnitude', 'deaths', 'tsunami', 'date'],
    color_discrete_sequence=['#cb0000'],
    projection="mollweide"
    )
main.add_trace(fig.data[0])
main.update_layout(width=1800, height=1000)
main.show()

url_cities = 'https://github.com/erikbozik/Earthquakes/raw/main/worldcities.csv'
cities = pd.read_csv(url_cities)

e_cop = earthquakes.copy(deep=True).query("`deaths`.notnull()")
e_cop['deaths'] = e_cop['deaths'].apply(deaths_parse)
e_cop.magnitude = e_cop.magnitude.astype(float)
e_cop['deaths'] = e_cop['deaths'].apply(lambda x: f'{x[0]} - {x[1]}' if x[1] != '>' else f'{x[0]}>')

main = px.choropleth(
    countries,
    locations=countries['iso_a3'],
    color_discrete_sequence=['#ffffff'],
    projection='mollweide',
    title='Mapa nízkej úmrtnosti v kombinácii s mestami po roku 1950'
)

cities_to_scatter = cities.copy(deep=True).query("`population` > 50000")
towns = px.scatter_geo(
    cities_to_scatter,
    lat=cities_to_scatter.lat,
    lon=cities_to_scatter.lng,
    hover_name='city',
    hover_data=['country', 'population'],
    projection='mollweide'
)

deaths_earthquakes = e_cop.copy(deep=True).query("`magnitude` > 7.5 and `deaths` == '1 - 50' and year > 1950")
deaths = px.scatter_geo(
    deaths_earthquakes,
    lat=deaths_earthquakes.geometry.y,
    lon=deaths_earthquakes.geometry.x,
    hover_name='date',
    hover_data=['deaths'],
    color_discrete_sequence=['#cb0000'],
    size='magnitude',
    projection='mollweide'
)

main.add_trace(towns.data[0])
main.add_trace(deaths.data[0])
main.update_layout(width=1800, height=1000, geo = dict(showlakes=False))
main.show()

e_high_deaths = earthquakes1.query("`deaths` == '1001>' and `year` > 1950")

main = px.choropleth(
    countries,
    locations=countries['iso_a3'],
    color='pop_dens',
    color_continuous_scale='Blues',
    range_color=(0, 250),
    projection='mollweide',
    title='Mapa vysokej úmrtnosti a vysokej hustoty obyvateľstva po roku 1950'
)
fig = px.scatter_geo(
    e_high_deaths,
    lat=e_high_deaths.geometry.y,
    lon=e_high_deaths.geometry.x,
    hover_name="location",
    hover_data=['magnitude', 'deaths', 'tsunami', 'date'],
    color_discrete_sequence=['#cb0000'],
    projection="mollweide"
    )
main.add_trace(fig.data[0])
main.update_layout(width=1800, height=1000)
main.show()

main = px.choropleth(
    countries,
    locations=countries['iso_a3'],
    color_discrete_sequence=['#ffffff'],
    projection='mollweide',
    title='Mapa vysokej úmrtnosti v kombinácii s mestami po roku 1950'
)

cities_to_scatter = cities.copy(deep=True).query("`population` > 50000")
towns = px.scatter_geo(
    cities_to_scatter,
    lat=cities_to_scatter.lat,
    lon=cities_to_scatter.lng,
    hover_name='city',
    hover_data=['country', 'population'],
    projection='mollweide'
)

deaths_earthquakes = e_cop.copy(deep=True).query("`deaths` == '1001>' and year > 1950")
deaths = px.scatter_geo(
    deaths_earthquakes,
    lat=deaths_earthquakes.geometry.y,
    lon=deaths_earthquakes.geometry.x,
    hover_name='date',
    hover_data=['deaths'],
    color_discrete_sequence=['#cb0000'],
    size='magnitude',
    projection='mollweide'
)

main.add_trace(towns.data[0])
main.add_trace(deaths.data[0])
main.update_layout(width=1800, height=1000, geo = dict(showlakes=False))
main.show()

most_e = earthquakes.copy(deep=True)
most_e['location'] = most_e['location'].astype(str)
def country(value):
    index = value.find(':')
    if index != -1:
        return value[: index]
    else:
        return value
most_e['source_country'] = most_e['location'].apply(country)
display(most_e)

counts_all_time = pd.DataFrame(most_e['source_country'].value_counts())
try:
  counts_all_time = counts_all_time.query('`count` > 23')
except:
  counts_all_time = counts_all_time.query('`source_country` > 23')
display(counts_all_time.shape)

(37, 1)

try:
  axes = sns.barplot(data=counts_all_time, x='count', y=counts_all_time.index)
except:
  axes = sns.barplot(data=counts_all_time, x='source_country', y=counts_all_time.index)
axes.figure.set_size_inches(15,10)

height = 0
for i, rect in enumerate(axes.patches):
    try:
        axes.text(rect.get_x() + rect.get_width() + 1, height, f'{counts_all_time["source_country"][i]:.0f}',
                  ha='left', va='center', fontsize=10)
        height += 1
    except KeyError:
        axes.text(rect.get_x() + rect.get_width() + 1, height, f'{counts_all_time["count"][i]:.0f}',
                  ha='left', va='center', fontsize=10)
        height += 1
axes.set_xlabel('Počet')
axes.set_title('Počet zemetrasení v jednotlivých štátoch')
pass

C:\Users\varga\AppData\Local\Temp\ipykernel_41628\1770827676.py:14: FutureWarning:

Series.__getitem__ treating keys as positions is deprecated. In a future version, integer keys will always be treated as labels (consistent with DataFrame behavior). To access a value by position, use `ser.iloc[pos]`

counts_last_hundreds = most_e.query("`year` > 1900")
counts_last_hundreds = pd.DataFrame(counts_last_hundreds['source_country'].value_counts()) # value counts rovno zosortuje
try:
  counts_last_hundreds = counts_last_hundreds.query('`count` >= 16')
except:
  counts_last_hundreds = counts_last_hundreds.query('`source_country` >= 16')
display(counts_last_hundreds.shape)

(37, 1)

try:
  axes = sns.barplot(data=counts_last_hundreds, x='count', y=counts_last_hundreds.index)
except:
  axes = sns.barplot(data=counts_last_hundreds, x='source_country', y=counts_last_hundreds.index)
axes.figure.set_size_inches(15,10)
height = 0
for i, rect in enumerate(axes.patches):
    try:
        axes.text(rect.get_x() + rect.get_width() + 1, height, f'{counts_last_hundreds["source_country"][i]:.0f}',
                  ha='left', va='center', fontsize=10)
        height += 1
    except KeyError:
        axes.text(rect.get_x() + rect.get_width() + 1, height, f'{counts_last_hundreds["count"][i]:.0f}',
                  ha='left', va='center', fontsize=10)
        height += 1
axes.set_xlabel('Počet')
axes.set_title('Počet zemetrasení od roku 1900')
pass

C:\Users\varga\AppData\Local\Temp\ipykernel_41628\3826353393.py:13: FutureWarning:

Series.__getitem__ treating keys as positions is deprecated. In a future version, integer keys will always be treated as labels (consistent with DataFrame behavior). To access a value by position, use `ser.iloc[pos]`

e_magnitude_trends = most_e.copy(deep=True).query("`year` > 1900 and `magnitude`.notnull()")
# display(e_magnitude_trends)
e_magnitude_trends['date'] = e_magnitude_trends['date'].apply(lambda x: pd.to_datetime(x, format='%Y/%m/%d', errors='coerce'))
e_magnitude_trends['magnitude'] = e_magnitude_trends['magnitude'].astype(float).apply(lambda x: round(x, 2))
magnitude_trends = counts_last_hundreds.copy(deep=True).iloc[0 : 10, :]
magnitude_trends = magnitude_trends.index

figure, axes = plt.subplots(figsize=(10, 10))
ax = sns.histplot(data=e_magnitude_trends.query("`source_country` in @magnitude_trends"), x='date', hue='source_country',
                  multiple='stack', bins=12)
ax.set_xlabel('Rok')
ax.set_ylabel('Počet zemetrasení')
ax.set_title('Distribúcia evidencie')
pass

figure, axes = plt.subplots(1, 2, figsize=(20, 10))
# print(len(axes))
ax = sns.kdeplot(data=e_magnitude_trends.query("`source_country` in @magnitude_trends"), x='date', hue='source_country', ax=axes[0])
# sns.kdeplot(data=e_magnitude_trends.query("`source_country` in @magnitude_trends"), x='date', hue='source_country', ax=axes[1])
ax.set_ylabel('Pravdepodobnosť (Density)')
# axes.axvline(x=pd.to_datetime('1940'), linestyle='--', color='black')
ax.set_xlabel('Rok')
ax.set_title('KDE evidencie')

ax = sns.kdeplot(data=e_magnitude_trends.query("`source_country` in @magnitude_trends"), x='date', hue='source_country', multiple='stack',
                 ax=axes[1])
ax.set_ylabel('Pravdepodobnosť (Density)')
ax.set_xlabel('Rok')
ax.set_title('KDE evidencie')
pass

url = 'https://github.com/erikbozik/Earthquakes/raw/main/PB2002_plates.json'
plates = gpd.read_file(url)

main = px.choropleth(
    plates,
    geojson=plates['geometry'],
    locations=plates.index,
    color=plates.index,
    color_continuous_scale='Brwnyl',
    projection='mollweide',
    title='Mapa zemetrasenia vzhľadom na tektonické dosky'
)
states = px.choropleth(
    countries,
    locations=countries['iso_a3'],
    color_discrete_sequence=['#ffffff'],
    projection='mollweide'
)

fig = px.scatter_geo(
    earthquakes,
    lat=earthquakes.geometry.y,
    lon=earthquakes.geometry.x,
    hover_name="location",
    hover_data=['magnitude', 'deaths', 'tsunami', 'date'],
    color_discrete_sequence=['#0043a5'],
    projection="mollweide"
)
main.add_trace(states.data[0])
main.add_trace(fig.data[0])
main.update_layout(width=1800, height=1000, coloraxis_showscale=False, geo = dict(showlakes=False))
main.show()

countries_counts_index = counts_last_hundreds.iloc[0: 10, :].index
e_counts = most_e.query("`source_country` in @countries_counts_index")

countries_counts_index = countries_counts_index.str.lower()


fig = px.scatter_geo(
    e_counts,
    lat=e_counts.geometry.y,
    lon=e_counts.geometry.x,
    hover_name="location",
    hover_data=['magnitude', 'deaths', 'tsunami', 'date'],
    color_discrete_sequence=['#0043a5'],
    projection="mollweide"
)

countries2 = countries.copy(deep=True)
countries2['name'] = countries2['name'].str.lower()
countries2 = countries2.query("`name` in @countries_counts_index")


main = px.choropleth(
    plates,
    geojson=plates['geometry'],
    locations=plates.index,
    color=plates.index,
    color_continuous_scale='Brwnyl',
    projection='mollweide',
    title='TOP 10 krajín v počte zemetrasení v porovnaní s tektonickými doskami'
)

states = px.choropleth(
    countries2,
    locations=countries2['iso_a3'],
    color_discrete_sequence=['#ffffff'],
    projection='mollweide'
)

main.add_trace(fig.data[0])
main.add_trace(states.data[0])

main.update_layout(width=1800, height=1000, geo = dict(showlakes=False), coloraxis_showscale=False)
main.show()

def gen_xy(rows, cols):
    def gen_x(rows, cols):
        for i in range(rows):
            for j in range(cols):
                yield i

    def gen_y(rows, cols):
        for i in range(rows):
            for j in range(cols):
                yield j
    return list(gen_x(rows, cols)), list(gen_y(rows, cols))

grid = (5, 2)
figure, axes = plt.subplots(*grid, figsize=(20, 25), sharex=True, sharey=True)
# for itering rows, columns
x, y = gen_xy(*grid)
# print(x, y)
for order, i in enumerate(magnitude_trends):
    data = e_magnitude_trends.query("`source_country` == @i")
    ax = sns.histplot(data=data, x='magnitude', ax=axes[x[order], y[order]], kde=True, stat='percent', bins=np.arange(3, 10, 0.2))
    ax.set_title(i)
    ax.tick_params(axis='both', which='both', length=10)
    ax.tick_params(labelbottom=True, labelleft=True)
    ax.set_xlabel('Magnitúda')
    ax.set_ylabel('Percento')
# plt.tick_params()
pass

figure, axes = plt.subplots(5, 2, figsize=(20, 25), sharex=True, sharey=True)

for order, i in enumerate(magnitude_trends):
    data = e_magnitude_trends.query("`source_country` == @i")
    ax = sns.lineplot(data=data, x='date', y='magnitude', ax=axes[x[order], y[order]])
    ax.tick_params(axis='both', which='both', length=10)
    ax.tick_params(labelbottom=True, labelleft=True)
    mean = data['magnitude'].mean()
    axes[x[order], y[order]].axhline(y=mean, color='black', linestyle='--')
    ax.set_title(f'{i}, Mean: {round(mean, 2)}')
    ax.set_ylabel('Magnitúda')
    ax.set_xlabel('Rok')
pass

amount = pd.DataFrame(volEvents.copy(deep=True)['morphology'].value_counts())
figure, axes = plt.subplots(figsize=(10, 10))
try:
  ax = sns.barplot(data=amount, y=amount.index, x='count')
except:
  ax = sns.barplot(data=amount, y=amount.index, x='morphology')
# ax.bar_label(ax.containers[0],)
height = 0
for i, rect in enumerate(axes.patches):
    try:
        axes.text(rect.get_x() + rect.get_width() + 1, height, f'{amount["morphology"][i]:.0f}',
                  ha='left', va='center', fontsize=10)
        height += 1
    except KeyError:
        axes.text(rect.get_x() + rect.get_width() + 1, height, f'{amount["count"][i]:.0f}',
                  ha='left', va='center', fontsize=10)
        height += 1

ax.set_xlabel('Počet')
ax.set_ylabel('Morfológia')
ax.set_title('Typy a počet výbuchov')
pass

C:\Users\varga\AppData\Local\Temp\ipykernel_41628\769557073.py:15: FutureWarning:

Series.__getitem__ treating keys as positions is deprecated. In a future version, integer keys will always be treated as labels (consistent with DataFrame behavior). To access a value by position, use `ser.iloc[pos]`

amount_VEI = volEvents.copy(deep=True)
amount_VEI.sort_values(by='explosivity_index_VEI', inplace=True)
amount_VEI['explosivity_index_VEI'] = amount_VEI['explosivity_index_VEI'].astype(str)
VEI_index ={6: 'colossal', 7: 'super-colossal', 5: 'paroxysmal',  0: 'non-explosive',  4:'cataclysmic',  3: 'severe',  2: 'explosive',  1: 'gentle'}
legend = ''.join([f'{i} ({VEI_index[i]})\n' for i in range(8)])
figure, axes = plt.subplots(figsize=(10, 10))
ax = sns.histplot(data=amount_VEI, x='explosivity_index_VEI', label=legend, color='brown', shrink=0.8)
ax.set_xlabel('Index vulkanickej aktivity')
ax.set_ylabel('Počet')
ax.set_title('Počet sopečných aktivít vzhľadom na VEI index')


for rect in ax.patches:
    height = rect.get_height()
    ax.annotate(f'{height:.0f}', xy=(rect.get_x() + rect.get_width() / 2, height), xytext=(0, 5),
                textcoords="offset points", ha='center', va='bottom', fontsize=10)
plt.legend()

<matplotlib.legend.Legend at 0x15900eba2d0>

VEI_range = volEvents.copy(deep=True)
try:
  to_display = amount.query('`count` >= 10').index
except:
  to_display = amount.query('`morphology` >= 10').index

VEI_range = VEI_range.query("`morphology` in @to_display")
figure, axes = plt.subplots(figsize=(10, 10))
ax = sns.boxplot(data=VEI_range, y='morphology', x='explosivity_index_VEI')
ax.set_xlabel('Index vulkanickej aktivity')
ax.set_ylabel('Morfológia')
ax.set_title('Rozsahy VEI indexov vzhľadom na morfológiu')
pass

figure, axes = plt.subplots(figsize=(10, 20))
ax = sns.violinplot(data=VEI_range, y='morphology', x='explosivity_index_VEI')
ax.set_xlabel('Index vulkanickej aktivity')
ax.set_ylabel('Morfológia')
ax.set_title('Rozsahy VEI indexov vzhľadom na morfológiu')
pass

deaths_distribution = volEvents.copy(deep=True)
figure, axes = plt.subplots(figsize=(10, 10))
ax = sns.histplot(data=deaths_distribution, x='deaths', bins=range(0, 11, 1), hue='explosivity_index_VEI', palette='rocket', multiple='stack', stat='percent')
ax.axhline(y=5)
axes.set_xticks(range(0, 16))
ax.set_xlabel('Smrte')
ax.set_ylabel('Percentuálne vyjadrenie')
ax.set_title('Distribúcia úmrtí pri vulkanickej aktivite')
pass

deaths_box = volEvents.copy(deep=True).query('`deaths` != 0')
deaths_box['explosivity_index_VEI'] = deaths_box['explosivity_index_VEI'].astype(float)
deaths_box['deaths'] = deaths_box['deaths'].astype(float)

figure, axes = plt.subplots(figsize=(10, 10))
ax = sns.scatterplot(data=deaths_box, x='explosivity_index_VEI', y='deaths')
for i in range(1000, 7001, 1000):
    plt.axhline(y=i, color='grey', linestyle='--')
ax.set_ylim(0, 7000)
ax.set_xlabel('Index vulkanickej aktivity')
ax.set_ylabel('Počet úmrtí')
ax.set_title('Závislosť úmrtí od VEI')
pass

figure, axes = plt.subplots(figsize=(10, 10))
deaths_hm = (deaths_box.copy(deep=True)[['explosivity_index_VEI', 'deaths']]
             .set_index('deaths')
             .sort_values(by='deaths', ascending=False))
deaths_hm['explosivity_index_VEI'] = deaths_hm['explosivity_index_VEI'].astype(int)
ax = sns.heatmap(data=deaths_hm, cmap='BuPu')
#ax.set_xlabel('Index vulkanickej aktivity')
ax.set_ylabel('Počet úmrtí')
ax.set_title('Závislosť úmrtí od VEI')
pass

# display(pd.unique(deaths_box['explosivity_index_VEI']))
deaths_box['explosivity_index_VEI'] = deaths_box['explosivity_index_VEI'].astype(int)
figure, axes = plt.subplots(figsize=(10, 10))
axes = sns.regplot(data=deaths_box, x='explosivity_index_VEI', y='deaths', x_jitter=True)
# nezobrazuje sa to dobre s log
# axes.set(yscale='log')
pearson = round(deaths_box['explosivity_index_VEI'].corr(deaths_box['deaths'], method='pearson'), 2)
spearman = round(deaths_box['explosivity_index_VEI'].corr(deaths_box['deaths'], method='spearman'), 2)
axes.set_title(f'Závislosť úmrtí od VEI\nPearson coeficient: {pearson}\nSpearman coeficient: {spearman}')
axes.set_xlabel('Index vulkanickej aktivity')
axes.set_ylabel('Počet úmrtí')
pass

# countries data
countries = gpd.read_file(gpd.datasets.get_path('naturalearth_lowres'))
country_areas = countries['geometry'].to_crs({'proj':'cea'}).area  / 1e6
countries['area'] = country_areas
density = (countries['pop_est']
           / countries['area'])
countries['pop_dens'] = density
display(countries.head())

C:\Users\varga\AppData\Local\Temp\ipykernel_41628\4169267502.py:2: FutureWarning:

The geopandas.dataset module is deprecated and will be removed in GeoPandas 1.0. You can get the original 'naturalearth_lowres' data from https://www.naturalearthdata.com/downloads/110m-cultural-vectors/.

v_trigger = volEvents.query('tsunami == True').copy(deep=True)
t_trigger = tsunamiEvents.loc[tsunamiEvents['cause'].str.contains('Volcano'), :].copy(deep=True)

t_trigger = t_trigger.rename(columns={
    'deaths': 't_deaths',
    'geometry': 't_geometry'
})

v_trigger = v_trigger.rename(columns={
    'deaths': 'v_deaths',
    'geometry': 'v_geometry'
})
# merging volEvents and tsunami
merged = v_trigger.loc[:, ['explosivity_index_VEI', 'v_deaths', 'date', 'v_geometry']]
def vol_date_to_year(x):
    if 'BC' in x:
        x = ('-' + x[: x.find('B')]).strip()
    return x
merged['date'] =  merged['date'].apply(vol_date_to_year)
tsunami_to_merge = t_trigger.loc[:, ['date', 't_deaths', 't_geometry', 'source_location', 'source_country']]
def tsunami_date_to_year(x):
    to = list(x[: x.find('/')])
    out = ''
    start = False
    for i in to:
        if not start:
            if i != '0':
                out += i
        else:
            out += 1
    return out
tsunami_to_merge['date'] = tsunami_to_merge['date'].apply(tsunami_date_to_year)
tsunami_to_merge.reset_index(inplace=True, drop=True)
merged.reset_index(inplace=True, drop=True)
merged = merged.merge(tsunami_to_merge, on='date', how='inner')
display(merged)

# viac ako tisic umrti
deaths = volEvents.copy(deep=True).query("`deaths` > 1000")
deaths['date'] = deaths['date'].apply(vol_date_to_year)
deaths['date'] = deaths['date'].astype(float)
deaths = deaths.query("`date` > 1950")

# deaths['deaths'] = deaths['deaths'].apply(vol_date_to_year).astype(float)
main = px.choropleth( # mapa ktora zobrazuje krajiny tu hustotu
    countries,
    locations=countries['iso_a3'], # prednaska
    color='pop_dens',
    color_continuous_scale='Blues', #paleta
    range_color=(0, 250), #rozsah
    projection='mollweide',
    title='Mapa úmrtí nad 1000 po roku 1950 vzhľadom na hustotu zaľudnenia'
)
fig = px.scatter_geo( # scatter na mape
    deaths,
    lat=deaths.geometry.y,
    lon=deaths.geometry.x,
    hover_name="name",
    hover_data=['explosivity_index_VEI', 'deaths', 'tsunami', 'date', 'morphology'],
    color_discrete_sequence=['red'], # pouzi jednu farbu
    size ='deaths',
    projection="mollweide",
    )
main.add_trace(fig.data[0]) # pridat do main fig z googelu
main.update_layout(width=1800, height=1000)
main.show()

deaths = volEvents.copy(deep=True).query("`deaths` <= 1000")
deaths['date'] = deaths['date'].apply(vol_date_to_year)
deaths['date'] = deaths['date'].astype(float)
deaths = deaths.query("`date` > 1950")
main = px.choropleth(
    countries,
    locations=countries['iso_a3'],
    color='pop_dens',
    color_continuous_scale='Blues',
    range_color=(0, 250),
    projection='mollweide',
    title='Mapa úmrtí pod 1000 po roku 1950 vzhľadom na hustotu zaľudnenia'
)

fig = px.scatter_geo(
    deaths,
    lat=deaths.geometry.y,
    lon=deaths.geometry.x,
    hover_name="name",
    hover_data=['explosivity_index_VEI', 'deaths', 'tsunami', 'date', 'morphology'],
    # color_continuous_scale='solar',
    # color='explosivity_index_VEI',
    color_discrete_sequence=['#cb0000'],
    size='deaths',
    projection="mollweide",
    )
main.add_trace(fig.data[0])
main.update_layout(width=1800, height=1000)
main.show()

main = px.choropleth(
    countries, # regiony
    locations=countries['iso_a3'],
    color_discrete_sequence=['#ffffff'],
    projection='mollweide',
    title='Mapa viac ako 1000 úmrtí, menej ako 1000 úmrtí vzhľadom na mestá'
)

cities_to_scatter = cities.copy(deep=True).query("`population` > 50000")
towns = px.scatter_geo(
    cities_to_scatter,
    lat=cities_to_scatter.lat,
    lon=cities_to_scatter.lng,
    hover_name='city',
    hover_data=['country', 'population'],
    projection='mollweide'
)

deaths_volcano = volEvents.copy(deep=True).query("`deaths` > 1000")
deaths = px.scatter_geo(
    deaths_volcano,
    lat=deaths_volcano.geometry.y,
    lon=deaths_volcano.geometry.x,
    hover_name='date',
    hover_data=['deaths'],
    color_discrete_sequence=['#cb0000'],
    size='explosivity_index_VEI',
    projection='mollweide'
)

low_deaths = volEvents.copy(deep=True).query("`deaths` <= 1000 and `deaths` != 0")
low = px.scatter_geo(
    low_deaths,
    lat=low_deaths.geometry.y,
    lon=low_deaths.geometry.x,
    hover_name='date',
    hover_data=['deaths'],
    color_discrete_sequence=['#00FF00'],
    size='explosivity_index_VEI',
    projection='mollweide'
)
main.add_trace(towns.data[0])
main.add_trace(deaths.data[0])
main.add_trace(low.data[0])
main.update_layout(width=1800, height=1000, geo = dict(showlakes=False))
main.show() # hover podla poradia pridanie

# lets do deaths compared to causes
tsunamiEvents_d = tsunamiEvents.query("`cause` != 'Unknown' and `deaths`.notnull()").copy(deep=True)
def deaths_parse(value):
    if value is not np.nan:
        index = value.find('(')
        key, value = value[: index].strip(), value[index + 2:]
        ran = ()
        for i in value.split():
             try:
                 ran += int(i),
             except ValueError:
                 pass
        if len(ran) != 2:
            out = (ran[0], 10000)
        else:
            out = (ran[0], ran[1])
        return out
    return value

def rek_deaths_parse(value):
    if value[1] == 10000:
        return f'{value[0]}>'
    return f'{value[0]} - {value[1]}'

tsunamiEvents_d['deaths'] = tsunamiEvents_d['deaths'].apply(deaths_parse)
tsunamiEvents_d.sort_values(by='deaths', inplace=True, ascending=False)
tsunamiEvents_d['deaths'] = tsunamiEvents_d['deaths'].apply(rek_deaths_parse)
# aby sa pomenovania xovej osi neprekryvali
tsunamiEvents_d['cause'] = tsunamiEvents_d['cause'].apply(lambda x: '\n'.join(x.split()))
figure, axes = plt.subplots(figsize=(15, 7))
axes = sns.histplot(data=tsunamiEvents_d, x='cause', hue='deaths', multiple='stack', element='bars', shrink=0.8, palette='GnBu')
axes.set_xlabel('Príčina tsunami')
axes.set_ylabel('Počet tsunami')
axes.set_title('Príčiny vzniku tsunami')
pass

tsunamiEvents_d2 = tsunamiEvents.query("`cause` != 'Unknown' and `deaths`.notnull()").copy(deep=True)
tsunamiEvents_d2 = tsunamiEvents_d2.query("`cause` not in ('Earthquake')")
tsunamiEvents_d2['deaths'] = tsunamiEvents_d2['deaths'].apply(deaths_parse)
tsunamiEvents_d2.sort_values(by='deaths', inplace=True, ascending=False)
tsunamiEvents_d2['deaths'] = tsunamiEvents_d2['deaths'].apply(rek_deaths_parse)
# tsunamiEvents_d2 = tsunamiEvents_d2.loc[~tsunamiEvents_d2['cause'].str.contains('Earthquake'), :]
figure, axes = plt.subplots(figsize=(15, 7))
axes = sns.histplot(data=tsunamiEvents_d2, x='cause', hue='deaths', multiple='stack', element='bars', shrink=0.8, palette='Oranges')
axes.set_xlabel('Príčina tsunami')
axes.set_ylabel('Počet tsunami')
axes.set_title('Príčiny vzniku tsunami')
pass

min_voltsunami = volEvents.copy(deep=True).query("`tsunami` == True")
min_voltsunami.sort_values(by='explosivity_index_VEI', inplace=True)
min_voltsunami['explosivity_index_VEI'] = min_voltsunami['explosivity_index_VEI'].astype(str)

VEI_index ={6: 'colossal', 7: 'super-colossal', 5: 'paroxysmal',  0: 'non-explosive',  4:'cataclysmic',  3: 'severe',  2: 'explosive',  1: 'gentle'}
legend = ''.join([f'{i} ({VEI_index[i]})\n' for i in range(8)])
figure, axes = plt.subplots(figsize=(10, 10))
ax = sns.histplot(data=min_voltsunami, x='explosivity_index_VEI', label=legend, color='brown', shrink=0.8)
for rect in ax.patches:
    height = rect.get_height()
    ax.annotate(f'{height:.0f}', xy=(rect.get_x() + rect.get_width() / 2, height), xytext=(0, 5),
                textcoords="offset points", ha='center', va='bottom', fontsize=10)
plt.legend()
ax.set_xlabel('Index vulkanickej aktivity')
ax.set_ylabel('Počet tsunami')
ax.set_title('Distribúcia VEI vzhľadom na tsunami')
pass

v_trigger = volEvents.query('tsunami == True').copy(deep=True)
t_trigger = tsunamiEvents.loc[tsunamiEvents['cause'].str.contains('Volcano'), :].copy(deep=True)

t_trigger = t_trigger.rename(columns={
    'deaths': 't_deaths',
    'geometry': 't_geometry'
})

v_trigger = v_trigger.rename(columns={
    'deaths': 'v_deaths',
    'geometry': 'v_geometry'
})
# merging volEvents and tsunami
merged = v_trigger.loc[:, ['explosivity_index_VEI', 'v_deaths', 'date', 'v_geometry']]
def vol_date_to_year(x):
    if 'BC' in x:
        x = ('-' + x[: x.find('B')]).strip()
    return x
merged['date'] =  merged['date'].apply(vol_date_to_year)
tsunami_to_merge = t_trigger.loc[:, ['date', 't_deaths', 't_geometry', 'source_location', 'source_country']]
def tsunami_date_to_year(x):
    to = list(x[: x.find('/')])
    out = ''
    start = False
    for i in to:
        if not start:
            if i != '0':
                out += i
        else:
            out += 1
    return out
tsunami_to_merge['date'] = tsunami_to_merge['date'].apply(tsunami_date_to_year)
tsunami_to_merge.reset_index(inplace=True, drop=True)
merged.reset_index(inplace=True, drop=True)
merged = merged.merge(tsunami_to_merge, on='date', how='inner')
display(merged)

# lets merge tsunami runups
runups_to_be_merged = tsunami_runups.copy(deep=True)
source_country = []
def loc_country(value):
  index = value.find(',')
  loc, country = value[:index].strip(), value[index + 1:].strip()
  source_country.append(country)
  return loc
runups_to_be_merged['source_location'] = runups_to_be_merged['source_location'].apply(loc_country)
runups_to_be_merged['source_country'] = pd.Series(source_country)
runups_to_be_merged['date'] = runups_to_be_merged['date'].apply(tsunami_date_to_year)
runups_to_be_merged.set_index(['source_country', 'date'], inplace=True)

runups_to_be_merged = runups_to_be_merged.reset_index().loc[:, ['geometry', 'source_location', 'source_country', 'date', 'max_height_m']]
merged_runups = merged.copy(deep=True).merge(runups_to_be_merged, on=['source_location', 'source_country', 'date'], how='inner')

display(merged_runups.head())

merged_runups = merged_runups.query("`max_height_m`.notnull()")
figure, axes = plt.subplots(figsize=(15, 10))
axes = sns.regplot(data=merged_runups, x='explosivity_index_VEI', y='max_height_m', x_jitter=True)
# axes.set_ylim(0, 20)
# axes.set(yscale='log')

pearson = round(merged_runups['explosivity_index_VEI'].corr(merged_runups['max_height_m'], method='pearson'), 2)
spearman = round(merged_runups['explosivity_index_VEI'].corr(merged_runups['max_height_m'], method='spearman'), 2)
axes.set_title(f'Výška tsunami v závislosti od VEI\nPearson coeficient: {pearson}\nSpearman coeficient: {spearman}')
axes.set_xlabel('Index vulkanickej aktivity')
axes.set_ylabel('Maximálna výška v metroch')
pass

# merging tables for visualisation
e_trigger = earthquakes.copy(deep=True)
e_trigger['tsunami'] = (e_trigger['tsunami'].astype(int)
             .map({
    1: True,
    0: False
}))
e_trigger = e_trigger.query("`tsunami` == True")
e_trigger['magnitude'] = e_trigger['magnitude'].astype(float)
e_tsunami = tsunamiEvents.loc[tsunamiEvents['cause'].str.contains('Earthquake'), :].copy(deep=True)

e_tsunami = e_tsunami.rename(columns={
    'deaths': 't_deaths',
    'geometry': 't_geometry'
})

e_trigger = e_trigger.rename(columns={
    'magnitude': 'e_magnitude',
    'deaths': 'e_deaths',
    'geometry': 'e_geometry'
})
# merging earthquakes and tsunami
merged = e_trigger.loc[:, ['e_magnitude', 'e_deaths', 'date', 'e_geometry']]
tsunami_to_merge = e_tsunami.loc[:, ['date', 't_deaths', 't_geometry', 'source_location', 'source_country']]

tsunami_to_merge.reset_index(inplace=True, drop=True)
merged.reset_index(inplace=True, drop=True)

merged = merged.merge(tsunami_to_merge, on='date', how='inner')
# merging tsunami runups
runups_to_be_merged = tsunami_runups.copy(deep=True)
source_country = []
def loc_country(value):
  index = value.find(',')
  loc, country = value[:index].strip(), value[index + 1:].strip()
  source_country.append(country)
  return loc
runups_to_be_merged['source_location'] = runups_to_be_merged['source_location'].apply(loc_country)
runups_to_be_merged['source_country'] = pd.Series(source_country)
runups_to_be_merged.set_index(['source_country', 'date'], inplace=True)
merged.set_index(['source_country', 'date'], inplace=True)

# mergin to show the whole range
runups_to_be_merged = runups_to_be_merged.reset_index().loc[:, ['geometry', 'source_location', 'source_country', 'date', 'max_height_m']]
merged_runups = merged.merge(runups_to_be_merged, on=['source_location', 'source_country', 'date'])

merged_runups.set_index(['source_country', 'date'], inplace=True)
merged_runups.rename(columns={
    'geometry': 'r_geometry',
}, inplace=True)
merged_runups.reset_index(inplace=True)
merged_runups = merged_runups.query("`e_magnitude`.notnull() and `max_height_m`.notnull()").sort_values(by='e_magnitude')
display(merged_runups)

figure, axes = plt.subplots(figsize=(15, 10))
axes = sns.regplot(data=merged_runups, x='e_magnitude', y='max_height_m', x_jitter=True)
axes.set_ylim(0, 20)
# axes.set(yscale='log')

pearson = round(merged_runups['e_magnitude'].corr(merged_runups['max_height_m'], method='pearson'), 2)
spearman = round(merged_runups['e_magnitude'].corr(merged_runups['max_height_m'], method='spearman'), 2)
axes.set_title(f'Magnitúda v závislosti s výškou tsunami\nPearson coeficient: {pearson}\nSpearman coeficient: {spearman}')
axes.set_xlabel('Magnitúda')
axes.set_ylabel('Maximálna výška v metroch')
pass

	pop_est	continent	name	iso_a3	gdp_md_est	geometry	area	pop_dens
0	889953.0	Oceania	Fiji	FJI	5496	MULTIPOLYGON (((180.00000 -16.06713, 180.00000...	1.928760e+04	46.141190
1	58005463.0	Africa	Tanzania	TZA	63177	POLYGON ((33.90371 -0.95000, 34.07262 -1.05982...	9.327793e+05	62.185621
2	603253.0	Africa	W. Sahara	ESH	907	POLYGON ((-8.66559 27.65643, -8.66512 27.58948...	9.666925e+04	6.240382
3	37589262.0	North America	Canada	CAN	1736425	MULTIPOLYGON (((-122.84000 49.00000, -122.9742...	1.003773e+07	3.744797
4	328239523.0	North America	United States of America	USA	21433226	MULTIPOLYGON (((-122.84000 49.00000, -120.0000...	9.509851e+06	34.515738

	location	magnitude	deaths	tsunami	date	year	geometry	source_country
0	JORDAN: BAB-A-DARAA,AL-KARAK	7.3	None	0	-2150/??/??	-2150	POINT (35.50000 31.10000)	JORDAN
1	TURKMENISTAN: W	7.1	Few (~1 to 50 people)	0	-2000/??/??	-2000	POINT (58.20000 38.00000)	TURKMENISTAN
2	SYRIA: UGARIT	None	Many (~101 to 1000 people)	1	-2000/??/??	-2000	POINT (35.80000 35.68300)	SYRIA
3	GREECE: THERA ISLAND (SANTORINI)	None	Many (~101 to 1000 people)	1	-1610/??/??	-1610	POINT (25.40000 36.40000)	GREECE
4	ISRAEL: ARIHA (JERICHO)	None	None	0	-1566/??/??	-1566	POINT (35.30000 31.50000)	ISRAEL
...	...	...	...	...	...	...	...	...
5879	CHILE	7.6	None	1	2016/12/25	2016	POINT (-73.88000 -43.41600)	CHILE
5880	FIJI ISLANDS	6.9	None	1	2017/01/03	2017	POINT (176.05300 -19.32600)	FIJI ISLANDS
5881	INDIA: AMBASA	5.5	Few (~1 to 50 people)	0	2017/01/03	2017	POINT (92.00600 24.01600)	INDIA
5882	IRAN: KHONJ	5.3	Few (~1 to 50 people)	0	2017/01/06	2017	POINT (53.11600 28.20500)	IRAN
5883	ITALY: FARINDOLA	5.7	Few (~1 to 50 people)	0	2017/01/18	2017	POINT (13.24100 42.60100)	ITALY

	pop_est	continent	name	iso_a3	gdp_md_est	geometry	area	pop_dens
0	889953.0	Oceania	Fiji	FJI	5496	MULTIPOLYGON (((180.00000 -16.06713, 180.00000...	1.928760e+04	46.141190
1	58005463.0	Africa	Tanzania	TZA	63177	POLYGON ((33.90371 -0.95000, 34.07262 -1.05982...	9.327793e+05	62.185621
2	603253.0	Africa	W. Sahara	ESH	907	POLYGON ((-8.66559 27.65643, -8.66512 27.58948...	9.666925e+04	6.240382
3	37589262.0	North America	Canada	CAN	1736425	MULTIPOLYGON (((-122.84000 49.00000, -122.9742...	1.003773e+07	3.744797
4	328239523.0	North America	United States of America	USA	21433226	MULTIPOLYGON (((-122.84000 49.00000, -120.0000...	9.509851e+06	34.515738

	explosivity_index_VEI	v_deaths	date	v_geometry	t_deaths	t_geometry	source_location	source_country
0	3	0	46	POINT (25.39600 36.40400)	None	POINT (25.39600 36.40400)	THERA ISLAND (SANTORINI)	GREECE
1	5	3500	79	POINT (14.42600 40.82100)	None	POINT (14.42600 40.82100)	BAY OF NAPLES	ITALY
2	0	0	416	POINT (105.42300 -6.10200)	Many (~101 to 1000 people)	POINT (105.42300 -6.10200)	JAVA-S. JAVA	INDONESIA
3	3	0	766	POINT (130.67000 31.58000)	Many (~101 to 1000 people)	POINT (130.70000 31.60000)	N. KAGOSHIMA BAY	JAPAN
4	0	0	1169	POINT (15.00400 37.73400)	None	POINT (15.00400 37.73400)	CALABRIAN ARC	ITALY
...	...	...	...	...	...	...	...	...
124	1	0	1983	POINT (123.59000 -8.54000)	None	POINT (123.59000 -8.54000)	LESSER SUNDA ISLANDS	INDONESIA
125	3	4	1994	POINT (152.20300 -4.27100)	None	POINT (152.21400 -4.23800)	PAPUA NEW GUINEA	PAPUA NEW GUINEA
126	2	0	1996	POINT (159.43000 54.05000)	None	POINT (159.43000 54.05000)	KARYMSKOYE, KAMCHATKA, RUSSIA	RUSSIA
127	0	0	1997	POINT (-62.18000 16.72000)	None	POINT (-62.18000 16.72000)	WHITE RIVER VALLEY	MONTSERRAT
128	3	0	1999	POINT (-62.18000 16.72000)	None	POINT (-62.18000 16.72200)	SOUFRIERE HILLS VOLCANO	MONTSERRAT

	explosivity_index_VEI	v_deaths	date	v_geometry	t_deaths	t_geometry	source_location	source_country
0	3	0	46	POINT (25.39600 36.40400)	None	POINT (25.39600 36.40400)	THERA ISLAND (SANTORINI)	GREECE
1	5	3500	79	POINT (14.42600 40.82100)	None	POINT (14.42600 40.82100)	BAY OF NAPLES	ITALY
2	0	0	416	POINT (105.42300 -6.10200)	Many (~101 to 1000 people)	POINT (105.42300 -6.10200)	JAVA-S. JAVA	INDONESIA
3	3	0	766	POINT (130.67000 31.58000)	Many (~101 to 1000 people)	POINT (130.70000 31.60000)	N. KAGOSHIMA BAY	JAPAN
4	0	0	1169	POINT (15.00400 37.73400)	None	POINT (15.00400 37.73400)	CALABRIAN ARC	ITALY
...	...	...	...	...	...	...	...	...
124	1	0	1983	POINT (123.59000 -8.54000)	None	POINT (123.59000 -8.54000)	LESSER SUNDA ISLANDS	INDONESIA
125	3	4	1994	POINT (152.20300 -4.27100)	None	POINT (152.21400 -4.23800)	PAPUA NEW GUINEA	PAPUA NEW GUINEA
126	2	0	1996	POINT (159.43000 54.05000)	None	POINT (159.43000 54.05000)	KARYMSKOYE, KAMCHATKA, RUSSIA	RUSSIA
127	0	0	1997	POINT (-62.18000 16.72000)	None	POINT (-62.18000 16.72000)	WHITE RIVER VALLEY	MONTSERRAT
128	3	0	1999	POINT (-62.18000 16.72000)	None	POINT (-62.18000 16.72200)	SOUFRIERE HILLS VOLCANO	MONTSERRAT

Úvod¶

Spracovanie raw dát¶

Načítanie dát¶

Analýza¶

1. Otázka: Závislosť magnitúdy zemetrasení a úmrtí¶

Magnitúda a úmrtia¶

Mapa nízkej úmrtnosti a nízkej hustoty obyvateľstva po roku 1950¶

Mapa nízkej úmrtnosti v kombinácii s mestami po roku 1950¶

Mapa vysokej úmrtnosti a vysokej hustoty obyvateľstva po roku 1950¶

Mapa vysokej úmrtnosti v kombinácii s mestami po roku 1950¶

2. Otázka: Kde a kedy sa zemetrasenia diali/dejú¶

Počet zemetrasení v jednotlivých štátoch¶

Počet zemetrasení od roku 1900¶

Distribúcia evidencie¶

KDE evidencie¶

Mapa zemetrasenia vzhľadom na tektonické dosky¶

TOP 10 krajín v počte zemetrasení v porovnaní s tektonickými doskami¶

Zastúpenie magnitúdy v TOP 10 krajinách¶

Vývin magnitúdy v čase¶

3. Otázka : Morfológie sopiek a výbuchy¶

Typy a počet výbuchov¶

Počet sopečných aktivít vzhľadom na VEI index¶

Rozsahy VEI indexov vzhľadom na morfológiu¶

4. otázka : Závislosť počtu smrtí od výbuchov¶

Distribúcia úmrtí pri vulkanickej aktivite¶

Závislosť úmrtí od VEI (scatterplot)¶

Závislosť úmrtí od VEI (heatmap)¶

Závislosť úmrtí od VEI pomocou Spearman-Pearson koeficientov¶

Úmrtia a hustota zaľudnenia¶

Mapa úmrtí nad 1000 po roku 1950 vzhľadom na hustotu zaľudnenia¶

Mapa úmrtí pod 1000 po roku 1950 vzhľadom na hustotu zaľudnenia¶

Mapa viac ako 1000 úmrtí, menej ako 1000 úmrtí vzhľadom na mestá¶

5. Otázka : Tsunami a ich prejavy¶

Príčiny tsunami a ich distribúcia vzhľadom na počet smrtí¶

Presnejší histogram s príčinami tsunami¶

Distribúcia VEI indexov vzhľadom na tsunami¶

Výška tsunami v závislosti od VEI¶

Magnitúda a výška tsunami¶

	source_country	date	e_magnitude	e_deaths	e_geometry	t_deaths	t_geometry	source_location	r_geometry	max_height_m
9249	CHINA	1992/01/05	3.7	None	POINT (108.00000 18.00000)	None	POINT (108.00000 18.00000)	SOUTH CHINA SEA	POINT Z (109.11000 21.02500 193.82003)	0.25
9251	CHINA	1992/01/05	3.7	None	POINT (108.00000 18.00000)	None	POINT (108.00000 18.00000)	SOUTH CHINA SEA	POINT Z (110.82000 19.57000 136.37172)	0.17
9252	CHINA	1992/01/05	3.7	None	POINT (108.00000 18.00000)	None	POINT (108.00000 18.00000)	SOUTH CHINA SEA	POINT Z (110.28083 20.02592 186.84337)	0.24
9253	CHINA	1992/01/05	3.7	None	POINT (108.00000 18.00000)	None	POINT (108.00000 18.00000)	SOUTH CHINA SEA	POINT Z (109.51000 18.23000 510.54501)	0.80
9250	CHINA	1992/01/05	3.7	None	POINT (108.00000 18.00000)	None	POINT (108.00000 18.00000)	SOUTH CHINA SEA	POINT Z (108.61667 19.10000 179.81022)	0.23
...	...	...	...	...	...	...	...	...	...	...
5373	CHILE	1960/05/22	9.5	Very Many (~1001 or more people)	POINT (-73.40700 -38.14300)	Very Many (~1001 or more people)	POINT (-73.40700 -38.14300)	SOUTHERN CHILE	POINT Z (140.45000 35.51700 1204.11998)	3.00
5374	CHILE	1960/05/22	9.5	Very Many (~1001 or more people)	POINT (-73.40700 -38.14300)	Very Many (~1001 or more people)	POINT (-73.40700 -38.14300)	SOUTHERN CHILE	POINT Z (139.82500 34.91890 688.78455)	1.21
5375	CHILE	1960/05/22	9.5	Very Many (~1001 or more people)	POINT (-73.40700 -38.14300)	Very Many (~1001 or more people)	POINT (-73.40700 -38.14300)	SOUTHERN CHILE	POINT Z (139.81700 34.91700 723.45567)	1.30
5378	CHILE	1960/05/22	9.5	Very Many (~1001 or more people)	POINT (-73.40700 -38.14300)	Very Many (~1001 or more people)	POINT (-73.40700 -38.14300)	SOUTHERN CHILE	POINT Z (140.40000 35.25000 1204.11998)	3.00
5695	CHILE	1960/05/22	9.5	Very Many (~1001 or more people)	POINT (-73.40700 -38.14300)	Very Many (~1001 or more people)	POINT (-73.40700 -38.14300)	SOUTHERN CHILE	POINT Z (141.08300 38.30000 1112.60500)	2.60